EPDs 101
by Larry Keenan, RAAA Director of Breed Improvement
One of the primary reasons for the existence of breed associations is to
provide tools that allow for improvement of the breed population, and
ultimately improve the reliability with which these breeds pass traits on
to commercial customers. These tools should be used in selection of
animals to remain in the population, and identification of animals to be
utilized in breeding decisions that will produce the next generation of
offspring. The Ranchers’ Guide to EPDs brochure provides an excellent
overview of RAAA EPDs; however, the balance of this article provides a
deeper look into the science of Animal Breeding and its application to
genetic progress.
Genetic Value vs.
Breeding Value
In order to increase the
competitive advantage of
the Red Angus product it is
imperative that we implement tools and strategies
that allow for genetic
improvement.
Genetic improvement necessitates the production of cattle
which express a superior phenotype for those traits that
have an impact on a producer’s profitability. In order to
create such animals we must
determine the proportion of an
animal’s total genetic value
that has the potential to be
transmitted to their offspring.
This genetic component is
referred to as an animal’s
breeding value. Due to the
fact that an animal’s breeding
value for any particular trait
cannot be directly measured,
we must calculate an estimated
breeding value (EBV). These
EBV calculations are based on
performance data of the ani-
42 American Red Angus Magazine n February 2009
mal and all related individuals.
EBVs are ideal in describing
the genetics of a particular animal; however, during sperm
and egg (gamete) production
an animal randomly contributes 50% of their genetics.
Therefore, in predicting the relative performance of progeny
it is more applicable to take
half of a parents EBV, which is
termed Expected Progeny
Difference (EPD). It is important to remember that numerous (sometimes hundreds)
genes, with varying impacts,
have an affect on traits important to beef production.
Therefore, an animal that has
an above average phenotype
for a given trait may possess
“good” genes, as well as “bad”
genes, for that particular trait.
During the process of random
sampling of genes during
EPDs 101
sperm/egg production, the
gamete may possess any number of possible combinations of
genes; therefore contributing
different impacts on a trait. It’s
due to this reality that increases the complexity of calculating EPDs.
RAAA National Cattle
Evaluation
EPD calculations are statistically based and become extremely
complex. In their most simplistic form an animal’s EPD is calculated by taking the average
of the sire and dam EPD. Often
referred to as a Parental EPD,
this genetic prediction is unreliable as it does not include
performance information on
the animal. Once performance
information is submitted to
RAAA, in-house EPD calculations are performed which
account for the performance of
the particular animal as well as
other animals in their contemporary group (a group of animals given equal opportunity
to perform). This EPD is
termed an Interim EPD, which
lacks the ability to fully
account for mating bias and
performance of related animals. These obstacles are overcome biannually when RAAA
performs a National Cattle
Evaluation (NCE), which is
performed by Colorado State
University Center for the
Genetic Evaluation of
Livestock (CSU-CGEL). At
44 American Red Angus Magazine n February 2009
CSU-CGEL, Best Linear
Unbiased Prediction (BLUP)
methodology is used to calculate EPDs. This methodology
has the ability to account for
environmental effects, mating
bias, and performance of all
related individuals in order to
predict an animals breeding
value. These calculations are
extremely advanced and can
not be simply calculated on a
calculator or an “off the shelf”
computer program.
An additional statistic calculated during the NCE is Accuracy
(ACC). ACC is a descriptor of
the reliability of an animal’s
predicted breeding value
(EPD) for the trait being evaluated. ACC ranges from 0 to 1
(0 = weak relationship, 1 =
strong relationship). For example, an animal that has an EPD
with a .99 ACC can be interpreted as meaning the EPD is a
very accurate prediction of the
animal’s true breeding value.
Conversely, an animal with a
.25 ACC should be understood
to mean the EPD is not an
accurate prediction of the animal’s true breeding value and
the EPD can change as more
data is collected.
Two major factors contributing
to ACC calculations: 1) amount
of performance data used in an
animals EPD calculation, and
2) heritability (strength of the
relationship between performance and breeding value of a
particular trait) of the trait
being evaluated. Just as in any
EPDs 101
prediction calculation, the more performance
data included directly results in a more accurate prediction. In addition, as heritability
increases the more useful each piece of performance data becomes in predicting the animals breeding value.
Genetic Change
The sole reason of using EPDs in breeding
decisions is to make the next generation of animals genetically better; therefore, having a better product to market. This concept is commonly referred to as genetic change. Genetic change
consists of ACC of EPDs used in selection,
selection intensity and variation of genetics in
the current herd. Selection intensity is simply
how demanding producers are in selecting animals to include in their breeding program. As
selection intensity increases, genetic change can
be made at a more rapid pace. The variability
of breeding values within a population can be
used to describe genetic variation. If genetic
variation is vast, future generations of progeny
will be better on average if producers only use
breeding animals that are superior for the trait
being selected on. The rate at which genetic
change can be made is dependent on generation interval – the length of time to replace one
generation with the next. As generation interval increases, the rate at which genetic change
can be made is reduced.
46 American Red Angus Magazine n February 2009
In order to increase the competitive advantage
of the Red Angus product it is imperative that
we implement tools and strategies that allow
for genetic improvement. Although producers
may use strategies in varying degrees to make
genetic change, it is critical that all producers
strive to improve the genetics of their herd.
Therefore, making the entire Red Angus population better, which will further increase the
demand for Red Angus cattle. n